Applying Evolutionary Multitasking for Process Parameter Optimization in Polymerization Process of Carbon Fiber Production

Carbon fiber is becoming a key material for engineering applications due to its excellent comprehensive properties. The process parameter optimization an important step in the polymerization of carbon production. At present, most research on usually carried out single production line, without considering correlation between problems. In this paper, multiobjective mechanism model co-optimization established. Each these submodels task, corresponding line. order solve effectively, we also designed evolutionary multitasking algorithm based transfer learning, which reuses past experiences one task generate population pool next iteration another enabling explicit genetic different tasks and accelerating convergence speed. proposed framework operation has been conducted 10 conditions process. Experimental results show that compared with other implicit algorithms, very competitive generating effective solutions. This provides support manufacturing production, will help engineers technicians make informed decisions.